This proposal extends previous research on psychosocial stressors and smoking cessation. The focus is on treating smokers who are recruited at the worksite and includes blue collar and lower SES groups. Relapse is still a major problem for treatment and, therefore, for reduction of coronary heart disease (CHD) risk. Individual difference profiles such as psychosocial stress, reactivity to smoking cues, and nicotine dependence are associated with "difficult" to treat smokers who are at highrisk for relapse. Previously, a standardized assessment of individual profiles was developed and evaluated. Next, a formative patient-treatment matching study was undertaken to evaluate a cognitive-behavioral treatment (CBT) tailored to one profile of risk for relapse. Results to date indicate two additional patient-treatment matching studies that warrant further research. The two studies focus on developing more cost-effective treatments, based on patient treatment matching, for the "difficult" to treat smoker at the worksite. Study 1 consists of a 2 x 2 randomized block design to investigate the relationship between stress/cue-reactivity (high vs. low responders) and CBT alone vs. CBT plus cue-exposure treatment. This study evaluates the incremental effectiveness of a cue-exposure relapse prevention strategy. The hypothesis is that cue-exposure treatment will improve treatment outcome in high cue-reactive smokers. Study 1 uses the ;most cost-effective treatment obtained in study 1 and evaluates the incremental effect of adding nicotine gum to the treatment. Subjects are stratified on the basis of high versus low scores on a nicotine dependence index in a 2 x 2 randomized block design. Study 2 rests the hypothesis that nicotine gum results in increased effectiveness for high nicotine dependent subgroups or smokers. This research is significant in that it builds a foundation for examining patient-treatment matching hypotheses to help imporve treatment outcomes for the more "difficult" to treat subgroups of smokers at the worksite. These smokers are unlikely to quit unaided or with minimal (e.g., self-help) treatments because they are more psychosocially, or biochemically dependent on cigarettes. Patient-treatment matching, if proved feasible, can result in substantially more cost-effective interventions at the worksite.